An Exploration of Factors Impacting Middle School Students’ Attitudes Toward Computer Programming

Abstract Computer programming is becoming an essential skill for young students regardless of their education or career goals. Therefore, for students to develop and for educators and researchers to accurately measure self-efficacy in and value for programming is important. Although student motivation in subject matter can be measured using self-report surveys, these types of instruments are prone to error due to inherent biases. In this quasi-experimental and cross-sectional study, we examined whether using a computer programming pretest before taking a perception survey (i.e., providing examples of the key concept in the survey beforehand) had an impact on students’ self-reported self-efficacy and utility value, thus helping collect more accurate data. Results showed no significant difference on self-efficacy and value scores between those who received a pretest and those who did not. In further analysis, however, it was found that pretest performance was positively correlated with self-efficacy and value. In addition, boys reported significantly higher self-efficacy and value than girls, confirming gender disparity outlined in previous research. An exploratory, but important, finding was that there was an interaction between gender and test performance for the self-efficacy and utility value. While boys’ who had high pretest scores also reported higher levels of self-efficacy and utility value, it was the opposite for girls.